Parallel Processing for Individual-Based Ecological Models

小行星9318160 绝大多数用于生态系统分析和管理问题的生态模型都依赖于种群和群落的聚集特性。 该奖项将利用高性能计算系统来分析基于组成这些种群和社区的许多个体生物之间相互作用的新模型。 这些基于个体的模型可以用来预测种群和群落的动态明确的景观，从而提供了一种方法来分析替代空间明确的管理方案对生态系统的影响。 大规模并行机提供了处理这些分析和管理问题的潜力，同时考虑到在更聚合的建模方法中被忽略的关于个体生物差异的大量信息。 该奖项涉及开发和比较替代方法，以在几个高性能计算系统上实施基于个人的模式，并将所获得的结果应用于生态系统分析和管理的几个具体问题，特别是几个物种之间的社区一级相互作用以及种群和社区的明确模型。 该奖项由生态学，生态系统，应用数学，计算数学和计算生物学项目共同支持。 ***

9318160 Gross The vast majority of ecological models applied to problems in ecosystem analysis and management have relied on aggregative properties of populations and communities. This award will take advantage of high performance computing systems to analyze new models based upon interactions between many individual organisms which comprise these populations and communities. These individual-based models may be used to predict the dynamics of populations and communities explicitly on a landscape, and thus provide a method to analyze the effect of alternative spatially explicit management schemes on ecosystems. Massively parallel machines offer the potential to deal with these analysis and management problems, as well as taking account of the vast array of information about individual organism difference that is ignored in more aggregated modeling approaches. This award involved the development and comparison of alternative methods to implement individual-based modes on several high performance computing systems and the application of results obtained to several specific problems of ecosystem analysis and management, in particular community-level interactions between several species and explicit models of populations and communities. This award is being jointly supported by programs in Ecology, Ecosystems, Applied Mathematics, Computational Mathematics and Computational Biology. ***